Machine Translation and Discriminative Models

Speaker:
Martin Popel
Abstract:
This talk will consist of two interweaving parts on the border of machine learning and machine translation. In the theoretical part I will focus on large-scale multiclass classification, online learning, label-dependent features and a choice of loss function. In the practical part I will report on my ongoing experiments with Vowpal Wabbit and deep-syntactic MT system TectoMT.
Length:
01:17:40
Date:
23/03/2015
views:

Images:
Preview of 1.jpg
Image 1.jpg
Preview of 2.jpg
Image 2.jpg
Preview of 3.jpg
Image 3.jpg
Attachments: (video, slides, etc.)
107 MB
169 downloads
1.4 GB
171 downloads
487 MB
161 downloads
248 MB
169 downloads
180 MB
160 downloads